From Data to Dialogue: How S&P Global Energy Made Its Structured Data Estate Conversational with Databricks Genie Agents and MCP
Databricks, Friday, September 25th, 2026
S&P Global Energy used Databricks Genie Agents exposed as managed MCP servers to make its structured data conversational for AI agents.
S&P Global Energy describes making its complex structured data estate, spanning chemicals, crude oil, refined products, gas and power, and LNG, available to AI agents through the Model Context Protocol.
Domain experts curate focused Genie Agents per dataset group without writing agent code, and these agents act as managed MCP servers composed via a FastMCP proxy into composite endpoints for cross-domain queries.
The approach preserved Unity Catalog governance while significantly shortening time-to-market for conversational data products, letting both customer and internal agents ask natural-language questions and get governed answers.